Frequently asked questions
What is the best voice for interviews?
Atlas suits news-style interviews with its neutral, professional tone, while Milo offers a friendly, conversational delivery. Both voices support all 23 languages, allowing you to maintain consistency across multilingual episodes without needing different voice talent for each segment. Select the preset that best matches your specific interview style.
Which voice is best for storytelling?
Luna is ideal for soft, storytelling narration, while Sage offers a calm, instructional tone. Since we don’t offer custom emotion tags, rely on these presets. You can adjust speed between 0.25 and 4.0 to match your narrative pacing. Note that any voice can speak any supported language, so ensure you set the language parameter for non-English text to avoid incorrect language guessing.
Can I use the same voice for all languages?
Yes, every voice can speak any of the 23 supported languages. You can use Aria for English, French, and German episodes, for example. Simply set the optional language parameter in your API request to ensure accurate pronunciation and accent for each specific language track.
How do I choose between Aria and Nova?
Choose Aria for a warm, clear narrator style, perfect for intros or calm segments. Choose Nova for a bright, upbeat energy, ideal for promotional spots or high-energy updates. Both are female voices with distinct tonal qualities, allowing you to vary the audio texture of your podcast.
What is the difference between tts-1 and tts-1-hd?
Both models use the same engine and have identical pricing. The difference lies in audio quality, with tts-1-hd offering higher fidelity. Use tts-1 for standard delivery and tts-1-hd when you need broadcast-ready clarity. Both support the same voices, formats, and speed settings for your API requests.
Are the voices natural sounding?
The voices are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobook readings. They sound natural and human-like. However, every generated file carries an inaudible watermark marking it as synthetic speech. They do not mimic specific celebrities or real people, ensuring a consistent AI-generated sound.
Can I change the voice mid-podcast?
Yes, you can change voices between requests. Since each API call generates a separate audio file, you can combine different voices in your editing software. Just split your script into parts, send each to the API with the desired voice parameter, and join the resulting audio files.
How do I handle different accents?
Set the optional language parameter to the ISO code of the target language. If omitted, Latin script defaults to English. For non-English Latin text, explicitly setting the language ensures correct accent and pronunciation. This applies to all 23 supported languages, from Arabic to Turkish, with any voice.
What is the best format for podcasts?
MP3 is the standard for podcast distribution, balancing quality and file size. WAV offers lossless quality for editing, while Opus and FLAC support efficient storage or high-fidelity needs. PCM is available for raw data processing. Since our API supports mp3, wav, opus, flac, and pcm, choose the format that best fits your specific workflow requirements.
How much does it cost to use multiple voices?
Voice selection does not affect the price. The cost is $0.01 per 1,000 characters regardless of which voice you use. You can mix Aria, Felix, and others in a single episode without any additional fees. The rate remains constant for both tts-1 and tts-1-hd models.
Can I use the voices commercially?
The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings. This generally permits commercial use, but you should verify specific license terms for your distribution channel. The API itself has no restrictions on commercial usage of the generated audio files.
Do the voices sound robotic?
The voices are based on real licensed recordings, providing a natural, human-like quality. They avoid the typical robotic monotone of older TTS systems. While they are synthetic and carry an inaudible watermark, the delivery is smooth and expressive, suitable for professional podcast production without sounding artificial.
How do I test voices before buying?
Sign in with Google to receive a $0.50 free trial credit, valid for 7 days. This covers about 50,000 characters, allowing you to test all 8 voices. You can generate samples using the API or the dashboard to hear each voice in your preferred language and format.
What is the best voice for news?
Atlas is our neutral, news-oriented voice, ideal for reporting. For a more energetic style, try Felix. Both support all 23 languages, ensuring consistent broadcast-quality audio for your segments. Note that "best" is subjective; choose based on whether you prefer a calm, authoritative tone or a brighter, upbeat delivery for your specific content needs.
Introduction to AI Voices
Text-to-speech technology converts written scripts into natural-sounding audio files via a simple API call. For freelance podcasters, this means turning rough drafts into broadcast-ready episodes without booking studio time or hiring voice talent for every segment.
What are AI Voices?
Our API generates speech using licensed recordings from LibriTTS-R and public-domain audiobooks. You send text and receive an audio file. We do not support SSML or custom lexicons. Each file carries an inaudible watermark to mark it as synthetic.
Benefits for Podcasters
Save time and money by generating voiceovers instantly. You get consistent audio quality across episodes, support for 23 languages, and full control over speed from 0.25x to 4.0x. Failed requests are free, so your budget stays predictable.
- No studio overhead or scheduling conflicts
- Multilingual support for global audiences
- Instant generation with transparent pricing
Choosing the Right Voice
Select from eight presets like Aria for warm narration or Felix for energetic promos. Match your voice to your content: use Luna for soft storytelling or Atlas for neutral news. Always set the language parameter for non-English text to ensure accurate pronunciation.
Female Voices Overview
Selecting the right female voice is critical for establishing your podcast’s tone and listener retention. Our three primary female presets—Aria, Nova, and Luna—offer distinct personalities tailored for different content styles, all powered by the same high-quality TTS engine.
Aria: Warm Narrator
Aria delivers a clear, warm narration style ideal for long-form storytelling or educational content. Her voice cuts through background music without sounding artificial, making her perfect for documentary-style episodes where clarity matters.
- Ideal for narrative storytelling
- Clear enunciation
- Professional yet approachable tone
Nova: Bright Upbeat
Nova brings an energetic, bright energy to your audio files. She excels in promotional segments, quick updates, or casual vlogs where a lively pace keeps listeners engaged. Her upbeat delivery feels natural and modern.
- High energy delivery
- Great for promos
- Engaging and lively
Luna: Soft Storytelling
Luna offers a soft, soothing voice perfect for bedtime stories, meditation guides, or intimate conversational podcasts. Her gentle pacing helps create a relaxing atmosphere, encouraging listeners to stay tuned for longer durations.
- Calming and gentle
- Soft pacing
- Best for relaxation content
Male Voices Overview
The male voices in PodcastVoice API offer distinct tonal profiles tailored for specific podcast formats. Whether you are producing high-energy promo clips or calm narrative segments, these four presets provide reliable, broadcast-ready audio without the need for voice actors or studio overhead.
Each voice is generated from licensed recordings, ensuring consistent quality and clear enunciation across 23 languages. You can adjust speed and format to match your exact production needs, from quick social media teasers to full-length episode intros.
Felix: Energetic Promo
Felix brings an energetic and promotional tone to your content. His voice is ideal for commercial breaks, sponsor reads, or dynamic show intros that need to capture attention immediately.
- Ideal for sponsor reads and ad breaks
- High energy suitable for promotional content
- Clear articulation for quick delivery
Milo: Friendly Conversational
Milo offers a friendly, conversational style that feels natural and approachable. He is well-suited for interview-style podcasts, casual storytelling, or any content where a relaxed, human-like tone is preferred.
- Natural, conversational delivery
- Great for interviews and casual talks
- Engaging and approachable tone
Atlas: Neutral News
Atlas provides a neutral, authoritative voice perfect for news updates, factual reporting, or educational content. His clear, unbiased tone ensures your information is delivered with credibility and precision.
- Neutral and authoritative tone
- Best for news and updates
- Clear, unbiased delivery
Orion: Deep Documentary
Orion’s deep, resonant voice is designed for documentary-style narratives and serious storytelling. He adds weight and gravity to your content, making him ideal for long-form narrative podcasts or historical accounts.
- Deep, resonant voice
- Perfect for documentaries and narratives
- Adds gravity to storytelling
Voice Selection by Genre
Choosing the right voice depends heavily on your podcast’s format. Interview shows thrive on natural, conversational tones, while narrative podcasts benefit from structured, engaging narration. News updates require clarity and neutrality to deliver facts without distraction. Selecting the correct preset ensures your content feels authentic and keeps listeners engaged.
Each genre has specific audio characteristics that align with our voice presets. By matching the voice to the content style, you create a consistent listening experience. This guide helps you navigate those choices for freelance podcasters.
Interview Shows
For interview-style podcasts, a friendly and conversational tone works best. Milo (Male) offers a warm, approachable voice perfect for hosting segments or introducing guests. His natural cadence mimics human interaction, making solo-hosted interviews feel dynamic.
Alternatively, Nova (Female) brings an upbeat energy that keeps conversations lively. Use her for quick recaps or segment transitions where you need to maintain high engagement without sounding robotic.
- Use Milo for host segments
- Use Nova for energetic transitions
Narrative Podcasts
Narrative podcasts require storytelling depth. Luna (Female) excels here with her soft, storytelling tone, ideal for emotional arcs or dramatic readings. Her voice adds warmth to scripted narratives.
For more structured storytelling, Aria (Female) provides a warm, clear narrator style. She is excellent for historical narratives or documentaries where clarity and emotional resonance are key.
Orion (Male) offers a deep, documentary-style voice. Use him for serious, immersive narratives that demand gravity and authority, such as investigative journalism or historical accounts.
- Luna for emotional stories
- Aria for clear narration
- Orion for deep documentaries
News and Updates
News podcasts demand neutrality and clarity. Atlas (Male) is designed for this purpose. His neutral tone delivers information without emotional bias, making it perfect for daily news briefs or update segments.
For promotional news or announcements, Felix (Male) adds an energetic touch. His upbeat style is ideal for highlighting key stories or driving listener action.
Ensure consistent delivery by sticking to one voice for news segments to maintain brand identity across episodes.
- Atlas for neutral news
- Felix for energetic updates
Language and Accent Considerations
Our API supports 23 languages, allowing you to reach global audiences without hiring multiple voice actors. You can mix languages in a single episode or create entirely localized content. The system guesses the language from the script, but you can explicitly set it for Latin-script text to ensure accuracy.
Hindi Voice Options
All eight voices can deliver natural Hindi audio. Since Hindi uses a distinct script, setting the language field ensures the engine applies the correct phonetic rules. This avoids the common issue of the API defaulting to English pronunciation for mixed scripts.
- Supports all 8 voices (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion)
- Explicit language code recommended for clarity
Spanish Voice Options
Spanish is fully supported across the entire voice library. Whether you need the warm narration of Aria or the energetic promo style of Felix, the audio quality remains consistent. This flexibility lets you maintain a unique brand identity even when speaking Spanish.
- Compatible with all 8 voice presets
- Consistent quality across different speaker types
Multilingual Content
You can switch languages between requests seamlessly. For podcasts with interviews in different languages, simply change the language parameter for each segment. The API handles the transition without requiring special formatting. Note that Vietnamese, Indonesian, and Thai are not yet supported.
- Switch languages per request
- No SSML required for language changes
- Unsupported: Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali
Technical Aspects
Technical precision matters when delivering broadcast-ready audio. Our API processes requests sequentially, meaning short phrases take a few seconds to generate, while longer texts approach real-time speed at roughly 25 characters per second. Under load, requests may queue, so plan your generation timelines accordingly.
Audio Quality
Both the tts-1 and tts-1-hd models use the same high-quality engine, ensuring consistent clarity across your episodes. The audio carries an inaudible watermark to identify it as synthetic speech, maintaining transparency for your audience.
Speed Control
You can adjust the playback speed between 0.25x and 4.0x using the speed parameter. This allows you to fine-tune the pacing of your podcast segments, making complex explanations easier to follow or speeding up ad reads.
Format Options
Choose from mp3, wav, opus, flac, or pcm formats depending on your editing workflow. Use mp3 for efficient web delivery, or select wav/flac for lossless quality during post-production editing.
Cost vs. Quality
Choosing between Standard and HD models affects audio fidelity but not your budget. The tts-1 model offers standard quality at the base rate, while tts-1-hd provides higher fidelity for the same price. Since both engines share identical pricing, you can upgrade specific episodes for broadcast polish without changing your cost structure.
Understanding character counting is crucial for accurate budgeting. Every space, punctuation mark, and character counts toward the 4096-character limit per request. This means dialogue-heavy scripts consume credits faster than narration-only content, so plan your splits carefully to avoid unexpected queue delays or wasted requests.
Standard vs HD
Both tts-1 and tts-1-hd use the same engine and cost $10 per 1 million characters. Use HD for your main show intro or key segments where audio quality matters most, and standard for filler content or social media clips.
The price difference is non-existent, making the choice purely about aesthetic preference. If your podcast features complex soundscapes, HD’s clarity might preserve more detail in the voice layer.
- Same price: $0.01 per 1,000 characters
- Identical engine performance
- Choose HD for intros/outros
Character Counting
Spaces and punctuation count as characters. A 4096-character limit means longer scripts must be split into multiple requests. Failed requests are free, so you can test splits without losing credit.
Roughly 25 characters generate one second of audio. For a 10-minute episode, you need about 150,000 characters, costing $1.50. Plan your text length to stay within API limits while maintaining narrative flow.
- Spaces count as characters
- 4096 chars max per request
- Failed requests are free
Budget Planning
With a free trial of $0.50, you can test voice selection before committing. Top-ups via USDT are minimum $5 and credit automatically. Balance does not expire, allowing you to build a reserve for seasonal content spikes.
No monthly plans or seats mean you only pay for what you generate. Track your usage via the dashboard to forecast costs for upcoming episodes or series launches.
- $0.50 free trial
- Min $5 USDT top-up
- No expiration on balance
Integration Examples
Integrating the PodcastVoice API into your workflow is straightforward, whether you prefer code or command-line tools. By leveraging the OpenAI-compatible endpoint structure, you can generate broadcast-ready audio files directly from your podcast scripts or content management systems.
The API accepts standard JSON bodies, allowing you to specify models, voices, and formats. This flexibility ensures that your generated audio fits seamlessly into your existing production pipeline, from drafting to final export.
Python Integration
Using the official OpenAI Python SDK is the most efficient way to automate voice generation. Simply configure the base_url to https://api.get28s.com/v1 and provide your API key. You can then call the Audio.speech method with your script text and desired voice preset.
This approach handles the request-response cycle cleanly, allowing you to save the resulting audio file directly to your local drive or upload it to your podcast hosting platform programmatically.
- Set
base_urltohttps://api.get28s.com/v1 - Use
Audio.speechwithmodel='tts-1' - Specify
voiceandinputtext - Save response as
.mp3or.wav
JavaScript Integration
For web-based podcast tools or browser extensions, the JavaScript SDK works identically to the Python version by changing the base_url. Update the baseURL to point to our API and use the same audio/speech endpoint. This enables real-time voice generation within your frontend applications.
You can integrate this directly into content creation dashboards, allowing users to preview voiceovers instantly before committing to a full episode generation.
Curl Examples
For quick testing or shell scripting, use curl to send a POST request to /v1/audio/speech. Include your Bearer token in the header and pass the JSON body with your text, voice, and format preferences.
This method is ideal for CI/CD pipelines or simple automation scripts where you need to generate audio files without installing a full SDK library.
curl -X POST https://api.get28s.com/v1/audio/speech-H "Authorization: Bearer tts-..."- `-d '{"model": "tts-1", "voice": "nova", "input": "Hello"}'`
Common Use Cases
AI voices enable podcasters to repurpose long-form content into bite-sized audio clips for social media without hiring additional voice talent. You can generate consistent, broadcast-ready clips from your script in minutes, maintaining brand identity across platforms.
This workflow saves time and budget, allowing solo creators to maintain a strong presence on platforms like Instagram, TikTok, and LinkedIn with minimal overhead.
Social Media Clips
Generate short, engaging audio snippets from your podcast episodes to drive traffic to your main show. Use different voices to highlight key quotes or teasers, keeping your audience engaged across multiple channels without recording new studio sessions.
- Repurpose episode highlights into 15-60 second clips
- Maintain consistent brand voice across platforms
- Quick turnaround for trending topics
Audiobooks
Turn written scripts or blog posts into listenable content. With support for up to 4096 characters per request and multiple languages, you can produce chapters in various languages. The synthetic watermark ensures clear attribution to your AI-generated content.
Ideal for creators who want to offer audio versions of their written work without the high cost of full studio narration.
- Support for 23 languages including Hindi and Spanish
- Clear synthetic watermark for attribution
- Cost-effective for long-form content
E-Learning
Create clear, instructional audio for online courses or training materials. Use Sage for calm, instructional tones or Felix for energetic promo-style announcements. Speed control (0.25–4.0) allows learners to adjust playback for better comprehension.
Ensure consistent audio quality across all modules, making your educational content accessible and professional without recurring studio fees.
- Adjustable speed for better comprehension
- Clear instructional tones like Sage
- Consistent quality across all modules
Troubleshooting Voice Issues
Even with high-quality voices, minor hiccups can occur during synthesis. Understanding common pitfalls helps you maintain professional audio standards without unnecessary delays or wasted credits.
This guide addresses three frequent issues: pronunciation quirks, tone mismatches, and consistency across segments. These are solvable with simple adjustments to your API requests.
Pronunciation
The engine guesses language based on script, defaulting to English for Latin text. If your content mixes languages or uses proper nouns, explicitly set the language parameter to avoid mispronunciations.
- Use the
languagefield for non-English Latin script text. - Split complex names into phonetic spellings if needed.
Tone Matching
Each voice has a distinct character. Aria suits warm narration, while Felix excels in energetic promos. If the tone feels off, switch voices rather than adjusting speed, which can distort the natural cadence.
- Match voice personality to podcast genre.
- Avoid extreme speed settings for natural tone preservation.
Consistency
Requests are processed sequentially, so load can cause slight delays. For long texts, split into 4096-character chunks to ensure stable generation. Use the same voice and speed settings across all segments to maintain a uniform audio experience.
- Keep
voiceandspeedconstant across segments. - Monitor queue times during high load.
Future of AI Voices
AI voice technology is advancing rapidly, offering podcasters ever more nuanced and natural-sounding options. While current models rely on licensed recordings, future iterations may incorporate deeper contextual understanding and emotional range without requiring complex manual tuning. This evolution promises to make synthetic voices indistinguishable from human ones for most broadcast purposes.
The shift toward more efficient models will likely reduce latency, making real-time podcasting and live interactions more viable. As infrastructure improves, the cost of high-fidelity audio generation is expected to drop further, democratizing access to professional-grade voiceovers for independent creators worldwide.
Technology Trends
The industry is moving toward more efficient neural networks that require less computational power. This allows for faster generation speeds and lower costs, enabling podcasters to produce content at scale without breaking the bank.
Multi-modal models that combine text, audio, and visual cues are emerging in the industry. While our API currently focuses on audio, future integrations could allow for synchronized video content creation, streamlining the podcast production workflow significantly.
- Efficient neural networks
- Multi-modal integration potential
Improvements
Expect better handling of complex punctuation and prosody, reducing the need for manual text adjustments. Voice cloning technologies are becoming more accessible, though we do not offer custom voice cloning.
Enhanced language support will continue to expand, allowing for more seamless multilingual content creation. Improvements in noise reduction and audio clarity will ensure that generated files are ready for broadcast with minimal post-processing.
What to Expect
Within the next few years, AI voices will become the standard for many podcast genres. Creators can expect faster turnaround times and lower costs, making high-quality audio accessible to everyone.
The integration of AI voices into popular podcast platforms will simplify distribution. While we focus on providing the API, the broader ecosystem will continue to evolve, offering more tools for seamless content creation and monetization.
Conclusion and Next Steps
The PodcastVoice API gives freelance podcasters a direct path to broadcast-ready audio without studio overhead. By leveraging eight distinct, licensed voices and 23 languages, you can produce high-quality voiceovers efficiently. The transparent pricing and simple API integration ensure that your creative workflow remains uninterrupted and cost-effective.
Start Your Project
Sign up on our /get-api-key/ page to receive your API key instantly. You can start with a $0.50 free trial valid for 7 days, allowing you to generate about 50,000 characters of speech. Use the provided curl examples or SDKs to test your first audio file immediately.
- Generate audio with tts-1 or tts-1-hd models
- Support for mp3, wav, opus, flac, and pcm formats
- Speed control from 0.25x to 4.0x
Explore More Voices
Choose from eight presets like Aria for warm narration or Orion for deep documentaries. Each voice is based on licensed recordings, ensuring high quality and legal clarity. You can mix and match voices to create dynamic episodes or maintain consistency across a series.
- Aria, Nova, Luna, Sage (Female)
- Felix, Milo, Atlas, Orion (Male)
- Support for 23 languages including Hindi and Spanish
Join the Community
Connect with other creators and stay updated on API changes via our contact page. We support the freelance podcasting community with reliable, transparent tools. Reach out via email for any questions about integration, billing, or voice selection.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. The price difference is non-existent, making the choice purely about aesthetic preference.
23 languages
Any voice speaks any of these languages.
